The overall objective of this study is to improve the diagnosis and staging of patients with prostate cancer through development and evaluation of more effective techniques for Doppler color flow imaging and quantification and of 3D correlation with histologic information. In a preliminary study, we have shown that frequency shift color flow Doppler ultrasound of the prostate had some success in defining the presence or absence of prostate cancer. This proposed study will qualitatively and quantitatively analyze in 2D and 3D the correlation of combined gray scale and dual color flow Doppler imaging features with surgical histology in patients with known prostate cancer and presumed normal glands. The proposed research will utilize state of the art gray scale imaging and power mode color Doppler imaging. We hypothesize that vascularity detected by power mode color Doppler imaging specifically, or in combination with the other two modes, relates to the presence of prostate cancer and to cancer tissue neovascularity, grade and stage, indicators of prostate cancer aggressiveness and clinical significance. Fifty (50) patients who will undergo radical prostatectomy, will be imaged and analyzed with 2D gray scale and color Doppler and new, 3D, gray scale and power mode and frequency shift color Doppler imaging techniques. The whole prostatectomy specimen will be marked with external landmarks, scanned with 3D gray scale ultrasound and sectioned for whole mounting in planes parallel to the gray scale in vitro scans. Regions of outstanding flow and gray scale properties will be marked in the in vivo image sets. Independently, regions of specific tissues will be outlined on the whole mount slides and digitized with the slides. The 3D specimen image set will be warped to coincide precisely with the 3D set of whole mount images and regions of interest, and the two in vivo image sets will be warped to match the in vitro set. Ultrasound measures of detected fractional blood volume, mean velocity and relative echogenicity will be measured in the volumetric regions of interest identified histologically and ultrasonically. The sonographically identified regions will be characterized histologically. Discriminant analysis will be performed to define a combination of ultrasound variables which best discriminates malignant from benign tissues and to then indicate the degree of correlation. A small group of patients undergoing radical cystoprostatectomy will serve as controls. The goal of this research is to further enhance our preliminary work and test the hypothesis that clinically significant prostate cancer is detectable with color Doppler imaging and separable from insignificant prostate cancer. We hope to further show that neovascularity is the histologic correlate to increased flow with color Doppler and that 3D imaging and power Doppler are techniques which can enhance the role of prostate ultrasound in the diagnosis and staging of prostate cancer.